Java 7 -> Java8 code breakage scenarios

[Boaz Nahum]

No. The crash case is much more complicated. I will re-produce it.

On Thu, Feb 28, 2013 at 2:24 PM, Maurizio Cimadamore <
maurizio.cimadamore at oracle.com> wrote:

> On 28/02/13 11:45, Boaz Nahum wrote:
>
>>   From your email I guessed you had to workaround new bugs. I probably
>>> have
>>> mis-interpreted. Good to know that everything is looking almost sane on
>>> your part.
>>>
>>>
>>> Yes, I hardy can read my own English ...
>>>
>> Actually, One of dependencies problem I reported may be consider as bug,
>> because the missing dependency cause a compiler crash. You helped me to
>> find it, see thread:
>>
>> http://mail.openjdk.java.net/**pipermail/compiler-dev/2013-**
>> February/005625.html<http://mail.openjdk.java.net/pipermail/compiler-dev/2013-February/005625.html>
>>
>> Once the dependency was added the crash disappeared.
>>
>> I will trace back all other issues, But it will take some time.
>>
> [moving to lambda-dev, which is more appropriate]
> I've tried to reproduce something like what you described:
>
> //A.java
> public interface A {
>     void m();
> }
>
> //B.java
> public class B implements A {
>     @Override
>     public void m() { }
>
>     public static void n() { }
> }
>
>
> //C.java
> public class C {
>     void test(B b) {
>         B.n();
>     }
> }
>
>
> I've compiled A, B, then only kept classfile for B.class.
>
> Then I tried to compile C.java against B.class - I only got an error
> message like this:
>
> C.java:3: error: cannot access A
>         B.n();
>          ^
>   class file for A not found
> 1 error
>
> The message goes away if you compile with -source 7. That's
> understandable, no default method lookup in JDK 7.
>
> However I could not reproduce any crash. So it seems like some ingredient
> is missing...
>
> Maurizio
>
>
>> Thanks
>> Boaz
>>
>
>